Holocene Occupation of the Mount Porr Strand Plain in Southern

NYAME AKUMA No. 76 DECEMBER 2011
chaeological materials observed during pedestrian
reconnaissance, therefore the presentation of data
in this manuscript is largely qualitative and oriented
toward showing the types of archaeological settlement of this portion of Lake Turkana. In this paper,
we present a classificatory and temporal scheme
for archaeological sites detected within our project
area as a basis for further study.
KENYA
Holocene Occupation of the
Mount Porr Strand Plain in
Southern Lake Turkana, Kenya
David K. Wright
Department of Archaeology and Art
History
College of Humanities
Seoul National University
Building 14-201
San 56-1, Sillim 9-dong, Gwanak-gu,
Seoul, 151-745
Republic of Korea
[email protected]
Background to the Study Area
Prior to the arrival of domesticated animals
in the Lake Turkana region, human subsistence was
focused on exploiting aquatic resources. This is
exemplified by the so-called “Bone Harpoon Cultures,” dated to ca. 8000 BP, but they likely span
additional millennia prior to and following this date
(Robbins 2006). All of these occupations were coeval to a presumed high stand of Lake Turkana of
+80m, or 455m asl (Owen et al. 1982). After 4000
BP, the subsistence economy shifted to cattle and
caprine herding complimented by ceramic use and
a microlithic tool industry called the “Pastoral Neolithic” (Bower and Nelson 1978). The period from
1000 until 200 BP can be defined as the “Turkana/
Maa/Cushitic” period following the evidence that
this was the period in which eastern Nilotic (proto-Turkana/Maa)-speaking people from southern
Sudan entered the Cushitic-speaking Turkana Basin (Ehret 2011: 117-119; Schlee 1985; Spear and
Waller 1993). Archaeological sites that date to
within the last 1000 years show great diversity in
the tools and modes of subsistence people practiced
(Robbins 1984). Figure 1 shows the spatial distribution of Holocene archaeological sites and hotspots
reported on the Turkana strand plain (Barthelme
1985; Beyin 2011; Butzer 1980; Phillipson 1977;
Robbins 1974, 1980).
Steven L. Forman
Department of Earth and Environmental Sciences
University of Illinois at Chicago
SES (m/c 186)
845 W. Taylor St.
Chicago, Illinois, 60607, USA
[email protected]
Introduction
This paper presents the results of two field
seasons of archaeological survey, mapping and limited test excavations of two sites adjacent to southeastern Lake Turkana. Within the study area, archaeological sites are located on small rises within
a ridge and swale landscape. The ridges and swales
are relics of past littoral beach processes. Evidence
for occupations spanning centuries to millennia are
located on these ridges, confirmed by radiometric
dating of stratified archaeological deposits. The
limited scope of these preliminary field investigations did not allow for complete collections of ar-
The present study area is known locally as
“Porr,” which refers to the 225m high cinder cone
volcano that dominates the visual scape of a thick
apron of Holocene-aged deposits northwest of the
town of Loiyangalani along the southeastern shores
of Lake Turkana (Figure 1). Holocene deposition
47
NYAME AKUMA across the Lake Turkana Basin is comprised of littoral and lacustrine deposits called the Galana Boi
Formation and rise to an elevation of 80m above the
modern lake level (Butzer 1980; Owen et al. 1982;
Owen and Renaut 1986). Beach ridge deposits are
often capped by eolian sediments, which reflect
No. 76 DECEMBER 2011
local reworking of medium-grained sands. Aggradation of eolian landforms occurs primarily in the
form of coppice dunes in which eolian sands accumulate around the margins of tree stands. Sandy alluvium filled drainages that dissect the project area
from north to south, exposing beach ridges and cop-
Figure 1: Archaeological sites and Digital Elevation Model (DEM) of the Lake Turkana Basin, surficial
processes and archaeological sites recorded within the project area. DEM provided by the United States
Geological Service (USGS) Earth Resources Observation and Science (EROS) Center (http://edc.usgs.gov/).
48
NYAME AKUMA pice dunes, and deposit sediments as alluvial fans
on the north side of a large east-west trending beach
ridge located at 11m above the modern lake level
(m a.l.l.).
No. 76 DECEMBER 2011
The perimeters of archaeological sites were
defined by the absence of archaeological artifacts
for at least 50m. For sites with smaller artifact assemblages (<20 artifacts), a total artifact collection
strategy was undertaken. For sites with >20 artifacts present on the ground surface, collection of
artifacts was generally made representatively, with
all identified temporally or culturally diagnostic artifacts piece-plotted and collected. In some cases,
archaeological sites were subdivided into loci, typically defined by landforms, when material classes
seemed similar and were spread over large areas.
All archaeological materials were returned to Nairobi for analysis and curation at the NMK.
Project goals and methods
Archaeological survey and limited test excavations were undertaken in 2008 on the Porr strand
plain of Lake Turkana (Figure 1). The project’s
goals were to map archaeological site occurrences
and recover a temporally representative dataset of
material culture. Future analyses will correlate the
archaeological data with geomorphic evidence of
the relic shorelines to reconstruct paleoecological
settlement systems from the middle to late Holocene. Our team plans to return with a total station
to map and collect artifact scatters on many of the
sites reported here.
Two archaeological sites identified in the
January survey were subjected to screened test excavations during the December 2008 field season.
Excavations were carried out primarily as 1m x 1m
test units (TUs) dug in 10cm spits and screened
through ¼” (5.25mm) mesh. One “stripping unit”
(SU) was excavated by removing the loose, sandy
overburden atop a buried soil to see if archaeological features could be identified in plan view from an
area of dense artifact scatter. Selection of the locations of test units was made judgmentally--typically in areas where dense artifact scatters were identified from surface contexts, but scattered across the
landform so as not to prejudice the identification of
stratified, subsurface deposits.
Archaeological survey was primarily conducted in January 2008, although political events
associated with a contested presidential election
curtailed full completion of the project objectives.
Nevertheless, a series of north-south transects were
walked with 15m interval spacing and recorded
using a Trimble XT handheld GPS unit, and postprocessed using Pathfinder 3.1 software. When archaeological artifacts were identified, a 5m grid was
established and artifacts were collected or recorded
with the GPS. The field crew included American
volunteers and staff from the National Museums
of Kenya (NMK) accompanied by a local “home
guard” to assist in navigation and translation with
local Rendille and Turkana people encountered
while doing fieldwork.
Description and interpretation of sediment
profiles used United States Department of Agriculture (USDA) nomenclature focusing on sediment
grain sizes, structure and soil properties (Soil Survey Staff 1993, 1999, 2002). All artifacts and fauna
were bagged in the field, returned to Nairobi for
photographing, morphometrics, and were curated
within the NMK Nairobi facility. Faunal analysis
was conducted by Paul Watene at the NMK and
included identification of faunal elements, bone
aspect, paleopathology (if present), presence of
burning on the artifacts, and species (when diagnostic attributes were present). Simuyu Wandibba
provided insight on the ceramic assemblages, and
Archaeological sites were recorded within
the Standardised African Site Enumeration System (SASES) accessioned at the NMK. In a few
instances, isolated lithic or ceramic artifacts were
identified and recorded as “sites” due to the lack
of a classification scheme of isolated occurrences
within the SASES system.
49
Balo
Adakurake
GeJk5
GeJk6
GeJk7
GeJk8
GeJk9
GeJk10
GeJk11
GeJk12
GeJk13
GeJk14
GeJk15
GeJk16
GeJk17
GeJk18
GeJk19
GeJk20
GeJk21
GeJk22
GeJk23
GeJk24
GeJk25
GeJk26
GeJk27
GeJk28
GeJk29
cairns
settlement/campsite
settlement/campsite
IO
scatter
scatter
ritual/ceremonial center
scatter
scatter
scatter
settlement/campsite
scatter
scatter (near manyatta)
scatter
settlement/campsite
scatter
settlement/campsite
campsite
settlement/campsite
IO
settlement/campsite
scatter
settlement/campsite
IO
IO
Type
* not in pedestrian survey corridor (located while driving)
X present, but not typologically diagnostic
NOTE: All sites with lithic tools also had flakes except IOs
Name
SASES
LSA - historic
LSA - PN/PIA
PN/PIA
unknown
LSA
PN/PIA
Rendille (historic)
PN/PIA
PN/PIA
PN/PIA
LSA - historic
PN/PIA
LSA - historic
PN/PIA
PN/PIA
LSA
LSA - historic
PN/PIA
PN/PIA
unknown
unknown
PN/PIA
LSA - PN/PIA
unknown
unknown
Cultural Identity
mammal
bovid, fish
Bovid 1, Bovid 2
X
Fauna
50
X
bovid, donkey, fish
ostrich egg
camel
X
ostrich egg
X, ostrich egg
bovid, fish, ostrich egg
Table 1: Archaeological sites identified in the 2008 field season.
NYAME AKUMA scrapers
scrapers, microliths, FCR
scrapers, FCR
scraper
scrapers, microliths
blade, scrapers, proj. point
scraper
scraper
scraper, FCR
scrapers
scrapers
blades, microliths
flakes
microliths, scrapers, FCR
1 flake
flaked cores, microliths
1 flake
1 flake, FCR
flaked cores, microliths
flaked core
Lithics
X
X
Nderit, X
Nderit, X
Nderit, X
corrugated, X
Nderit, corrugated, X
Nderit, corrugated, X
large, grit-temper
X
X
Nderit, X
Nderit, X
handled vessel frags, X
Nderit, X
handled vessel frags, X
large, grit-temper
X
Ceramics
No. 76 DECEMBER 2011
18.23
6.3
0.59
0.01
0.23
4.03
0.03
0.06
0.24
0.31
1.26
0.16
0.1
0.16
0.15
0.01
3.18
0.08
3.28
0.01
0.07
0.12
0.46
0.01
0.01
size (ha)
NYAME AKUMA Charles Nelson’s detailed photographs and descriptions of Turkana ceramics were also consulted for
further clarification of diagnostic wares (http://
www.chaz.org/). Generalized stone tool analysis
was performed by DKW following Nelson’s (1973)
and Mehlman’s (1989) typologies. Flaking, reduction patterns and raw material were also noted with
the assistance of Sonia Harmand.
No. 76 DECEMBER 2011
“Settlements and/or campsites” typically consisted
of numerous ceramic sherds and lithic flakes scattered on gentle ridges and knolls that characterize
the project area. One site (GeJk9) had evidence for
cooking activities (ceramic vessel fragments and
faunal material), but was only 0.08ha in area so
it was provisionally accessioned as a campsite. At
sites GeJk5 and GeJk28, hearthstones were identified intermixed with LSA ceramics and lithics. An
OSL age of beach ridge deposits underlying GeJk28
provides a maximum constraining age of settlement
at 6530±465 BP (UIC2719). GeJk28 is located at
+75m relative to the 2008 lake level (445-446 m asl
on the SRTM DEM using the EGM96 geoid). Artifact densities at GeJk5 and GeJk28 were relatively
high compared to the other archaeological sites
documented in the project area. A total of 60 lithic
and ceramic artifacts were collected from GeJk5
and 38 shell, lithic and ceramic artifacts were collected from GeJk28, but hundreds more artifacts
were observed at both sites and will be piece plotted
at a later date.
Results of Archaeological Investigations
Cultural-temporal categories. Archaeological sites were assigned five primary temporal/
cultural components: LSA, PN/Pastoral Iron Age
(PIA), historic and LSA through historic. The first
category includes Mode 5 microliths (Clark 1977)
made from obsidian, chalcedony, silcrete, chert or
quartz flakes. Two of the sites also included large basalt or phonolite reduced cores, which are common
from middle Holocene assemblages (Barthelme
1977). In our classificatory scheme, these sites
lack ceramics to distinguish them from ceramicyielding sites, and are located above +75m, which
is close to the Holocene transgressive limit of the
lake. Ceramics were recovered from PN/PIA sites,
which also typically had microliths, but there was
no clear evidence of historic habitation. PIA and PN
sites were not differentiated in our study because
of the difficulty in defining the “PIA” in the Lake
Turkana context. Historic sites have exogenouslyintroduced artifacts such as glass, metal alloys, or
plastic. The latter category generally includes archaeological assemblages with prehistoric artifacts
that are unattributable to a specific archaeological
culture. There were three settlement/camps that included historic surface components and one site,
Adakurake (GeJk13) that had discrete loci where
the differing ratios of historic vs. prehistoric artifacts were discernable. There were also four sites
categorized as having “unknown” components because the surface deposits were too sparse to assign
a site classification.
Archaeological sites that occur further below the high water mark of Lake Turkana have
low-density artifact scatters, typically consisting
of 20 or fewer artifacts. Notable caveats are the archaeological sites of GeJk10 and GeJk13, which
were subject to test excavations and described in
the section below. Hearthstones were provisionally
identified at GeJk13 and GeJk18, but were associated with decomposed manyatta (residences for
pastoralists made from animal skins strewn atop a
wooden frame). Heavily fragmented faunal remains
and charcoal were recovered from a shovel test at
GeJk14, indicating cooking activities occurred.
However, in most cases, the term “campsite” likely
best represents the character of the archaeological
sites rather than “settlement.” This is because the
archaeological assemblages tended to be diverse
(including more than one material class), but indicated only ephemeral human settlement.
Pedestrian reconnaissance. Archaeological surveys identified 21 archaeological sites that
were divided into five different categories (Table 1).
The other primary types of archaeological
sites identified were “scatters” and “IOs” (isolated
occurrences). Scatters are defined as having mul-
51
NYAME AKUMA tiple artifact finds, but lacking clear evidence for
cooking or settlement activities. In most cases, <10
artifacts were identified and recovered, although in
the case of GeJk17, 24 individual artifacts were distributed over 4.03ha (surveying the site in 5m intervals). IOs are single artifact finds recovered from
the ground surface. The surveys in 2008 identified
four IOs and nine scatters.
No. 76 DECEMBER 2011
and sheep were corralled into the ceremonial center
and forced to pass through two ends of an opening
between which fires were lit. The goats were then
sent to pasture. This ceremony did not necessarily
occur at the same location every year.
Lake Turkana is rimmed with so-called
“stone pillar sites,” which are interpreted as ceremonial centers related to mortuary practices (Hildebrand et al. 2011; Nelson 1995; Robbins 2006) or
astronomical markers (Lynch and Robbins 1978).
The scale of the linear stone features at GeJk18 is
far smaller than those of Jarigole or other stone pillar sites. Stone pillars can measure as much as 2.5m
long and are frequently associated with human remains. Neither scenario applies to GeJk18.
One site, GeJk18, was interpreted as a ritual/ceremonial center based on information provided
by local Rendille informants. GeJk18 consisted of
a line of rocks, some of which appear to be stacked
on top of each other and are generally oriented in a
straight line (Figure 2). The stones measured generally less than 20cm along the longest axis. The
project’s guard/informant (“Anton”) identified this
site as a place where Rendille met annually to exchange livestock and brides, and feast together at
some point in the past (zamani). In the rituals, goats
The final site category defined in the survey
was cairns. Our Rendille informant indicated that
the cairns were graves, but we did not attempt to
Figure 2: GeJk18.
52
NYAME AKUMA Figure 3: Balo (GeJk10) site map and representative profile.
Figure 4: Adakurake (GeJk13) site map and representative profile.
53
No. 76 DECEMBER 2011
NYAME AKUMA excavate any of the cairns as confirmation. Isolated
cairns were not given site numbers, but were located by GPS or were incorporated within other site
boundaries. However, one location, GeJk29, had a
cluster of 39 cairns located on a high volcanic ridge
adjacent to Lake Turkana. The locations of cairns
were recorded and artifacts were collected from the
ground surface. Most of the artifacts identified are
historic or modern in origin. However, three nondecorated potsherds, a nose scraper and an endscraper collected from the ground surface are interpreted as likely dating from the PN/PIA occupation
of the area.
No. 76 DECEMBER 2011
Balo to expose subsoil that corresponds to the AB
horizon in an area of dense artifact scatter, and to
identify features related to past human occupation.
However, no cultural features were identified in either the test or stripping units at either site.
In situ artifact deposits were primarily concentrated in the upper 20cm of each test unit excavated, although deeper archaeological deposits
at Adakurake reflect repeated occupation of the
landform during sediment accretion. Coppice dune
accretion was the mechanism of burial for cultural strata at Adakurake and appears to have been a
continuous process through human occupation of
the site. Ostrich eggshell from Adakurake recovered 40-50cm below the modern ground surface in
TU2 provided a 14C age of 2575±20 14C years BP
calibrated to 2725-2745 BP (ISGS A-1341; Reimer
et al. 2010). This level is the earliest detected occupation. At Balo, an OSL age of 2555 ± 175 BP
(UIC2718) derived from quartz grains assayed from
(C-horizon) littoral sediments provides a close, but
possibly maximum limiting age for the site.
Archaeological excavations and surface collections. Archaeological excavations and intensive
surface collections were undertaken at two sites and
demonstrated that settlement within the strand plain
was ephemeral. At Balo (GeJk10) a total of six 1m2
test units were screened and excavated to culturally sterile levels (Figure 3), whereas at Adakurake
(GeJk13) 8.5m2 were excavated (Figure 4). A 7.8m
x 11.9m stripping unit (SU) was also excavated at
Table 2: Fauna recovered from Balo (GeJk10).
TU
shovel test
surface
surface
surface
surface
surface
surface
surface
1
3
3
3
4
5
5
Taxonomy
fish
adult bovid
adult bovid
mammal
mammal
bovid
mammal
terrestrial
terrestrial
terrestrial
terrestrial
mammal
terrestrial
terrestrial
terrestrial
Element
spines
Px metacarpal
R ischium
long bone
bone
tooth
tooth
bone
bone
bone
bone
tooth
bone
bone
bone
Burned?
Size
2
1
2
X
X
54
#
2
1
1
1
2
1
1
7
3
33
1
2
32
2
1
NYAME AKUMA No. 76 DECEMBER 2011
Table 3: Fauna recovered from Adakurake (GeJk13).
TU
Taxonomy
Element
Burned?
Size
#
surface
adult bovid
phalanx 3
X
3
1
surface
bovid
R humerus
X
3
1
surface
mammal
cervical
X
1
surface
mammal
lumbar
X
1
surface
mammal
vertebra
X
2
surface
mammal
bone
1
bovid
tooth
1
bovid
tooth
14
1
fish
exonoste
1
1
fish
vertebra
1
fish
vertebra
1
terrestrial
bone
1
terrestrial
bone
1
mammal
centrum
X
1
1
mammal
petrosal
X
1
2
bovid
tooth
2
bovid
tooth
2
2
ostrich
eggshell
4
2
terrestrial
bone
5
3
terrestrial
bone
2
4
bovid
tooth
4
ostrich
eggshell
4
terrestrial
bone
4
terrestrial
bone
10
5
ostrich
eggshell
9
5
terrestrial
bone
18
6
terrestrial
bone
50
6
tilapia
3
6
tilapia
16
8
terrestrial
spines
spines/
vertebrae
bone
9
ostrich
eggshell
1
9
terrestrial
bone
2
2
X
2
5
2
X
139
128
2
4
4
1
1
X
4
2
55
1
NYAME AKUMA The fauna recovered in situ was fragmentary but derived from terrestrial bovid(s), fish and
ostrich eggshells (Tables 2 and 3). Adakurake had
the most diverse faunal assemblage of the two sites
tested, and shows that people were engaging in
mixed fishing, foraging and likely pastoral subsistence. Boscia coriacea and Salvadora persica trees
grow in abundance at Adakurake, which are important food sources for pastoralists who inhabit the
Turkana region (Stave et al. 2007). Donkey (Equus
asinus) and fish fragments were recovered from the
ground surface of GeJk17 and a camel astragalus
was recovered from the ground surface of GeJk21.
Ceramics recovered from the two sites consisted mostly of undecorated sherds (87% at Adakurake [n=26], 82% at Balo [n=102]), but decorated sherds were identified as variants of the Nderit
ceramic tradition (Figure 5). Nderit tradition ceramics recovered from both excavations and pedestrian
reconnaissance were typically thin-walled sherds
with pocket evulsions stippled into the interior of
No. 76 DECEMBER 2011
the vessel. Some Nderit tradition sherds also had
trellis rims and cross-hatching patterns on the body
of the sherd, and two rippled body sherds were also
recovered. Large, low-fired, grit-tempered sherds
were recovered from GeJk5 and GeJk23 (Figure
5) and may be fragments of figurines rather than
vessels based on their shapes. Three other sites
(GeJk20, GeJk21, and GeJk22) had well-fired, corrugated ceramic sherds (Figure 5) that were originally accessioned as Turkwell tradition ceramics
(cf. Lynch and Robbins 1979). However, based on
these sherds’ association with Nderit tradition ceramics, lack of clear decorative association and the
absence of material cultural attributes typically associated with Turkwell artifact assemblages such as
fishing implements, these sherds are classified as
“corrugated” pending further study.
Lithics fall into three primary categories:
reduced basalt or phonolite cores and flake tools;
Mode 5 microliths and flakes produced from quartz,
chert, chalcedony, silcrete and obsidian (Figures 6
Figure 5: Ceramics recovered from the 2008 project area.
56
NYAME AKUMA No. 76 DECEMBER 2011
Figure 6: Bifaces, burins and crescents recovered from the 2008 project area.
and 7); and fire-cracked rock (FCR). The centripetally-reduced igneous cores are morphologically
indistinguishable from Mode 3 tools (Clark 1977),
but are clearly associated with Mode 5 LSA/PN/
PIA assemblages. Obsidian blades measuring
3-4cm in length and scraper microliths recovered
from GeJk17 and GeJk26 are common features of
middle Holocene sites elsewhere in the Turkana
Basin (Barthelme 1977). Overall, the lithic assemblages from the Porr project area are geared
toward chopping and splitting (large cores), hunting (projectile points/bifaces), cutting soft or fleshy
items (blades and crescents), boring (burins) and
scraping hides or scales (scrapers). The assemblages are not specifically geared toward procuring fish, intensive cultivation or processing plants.
57
NYAME AKUMA No. 76 DECEMBER 2011
Figure 7: Microlithic scrapers and blades recovered from the 2008 project area.
Discussion
Phillipson 1977, 1984; Robbins 1972) that the artifact density is higher in middle Holocene sites than
compared to the later prehistoric and historic sites.
Our research goals during the 2008 field
season included developing a cultural historical
framework from which future studies could stem.
Sites GeJk5, GeJk26 and GeJk28 are the most likely candidates to have pre- or early-pastoral settlement components based on their locations near the
limit of the late Holocene transgressive strand plain
(>75m above the modern lake level) and dense
LSA microlithic artifact assemblages. The absence
of bone harpoons identified during pedestrian reconnaissance differentiates these sites from other
middle Holocene settlements in the region. Site
GeJk5 had coarse grit-tempered, thick-walled pottery sherds, large basalt cores, and at least one biface fragment that was likely used as a projectile
point—attributes of which signify that it might be
an early PN site (e.g., Barthelme 1977). Five obsidian blades, three burins (two obsidian) and two
obsidian scrapers were recovered from GeJk26,
which are highly characteristic of middle Holocene LSA assemblages. The 2008 survey results
agreed with previous surveys (Barthelme 1985;
Direct evidence for the presence of domesticated animals within the 2008 project area was
scant. Forelimb bones from a domesticated donkey
(Equus asinus) were recovered at GeJk17 and an
astragalus of a camel (Camelus dromedarius) was
recovered at GeJk21. But, the provenance of surface finds is difficult to assess, and the absence of
diagnostic faunal material from the ground surface
is interpreted as a preservation bias. Previous studies of archaeological sites across the prograded
strand plains of the Turkana Basin show middle
to late Holocene subsistence strategies involved
keeping domesticated livestock and engaging in
lakeside foraging (Barthelme 1985; Marshall et
al. 1984; Robbins 1984). Fauna from Adakurake
(GeJk13) show that people exploited both terrestrial (ostrich eggs, bovids) and aquatic (tilapia)
resources. Although the evidence is limited at the
Balo site (GeJk10), the fauna remains consist of
terrestrial and limited aquatic genera. Adult bovids
58
NYAME AKUMA in size classes 1 and 3 (Brain 1981) in the faunal
assemblages of Balo and Adakurake indicate that
people were hunting, while the bovids from size
classes 2 and 4 are potentially domesticated animals (ovicaprids and cattle, respectively). Riparian
plant species from the nearby drainages would have
been an important component of the subsistence regime of people inhabiting this landform (Morgan
1981; Stave et al. 2007), but direct evidence from
archaeological remains has yet to be identified.
No. 76 DECEMBER 2011
even if the archaeological assemblages do not include clear evidence for sustained PN occupations
or bone harpoon “cultures” found elsewhere in the
Lake Turkana region.
Acknowledgements
This research was funded by the National
Geographic Society, Committee for Research and
Exploration grants #8142-06 and 8456-08. The authors thank the Society for their continuing patience
as we struggled with logistical and technical obstacles in completing our research. The research was
conducted with permission from the Office of the
President of the Republic of Kenya (permit MOHEST 13/001/30C 220) and in collaboration with
the National Museums of Kenya Department of
Earth Sciences. Special thanks goes to Purity Kiura,
head of the Archaeology Section, for her invaluable
assistance and advice in facilitating the fieldwork
and laboratory analyses. The British Institute in
Eastern Africa loaned us field supplies and provided
logistical support. We would like to thank our field
assistants including Tom Mukhuyu, John Mwangi
and the late Dominic Sila of the National Museums
of Kenya, Boniface Ngira of the University of Nairobi, John Myers, Sean Stretton of the University of
Illinois, as well as Loiyangalani residents Antone,
Wilson, James, Yasas, Dekwa, and Lokom. Jim
Teasdale of the African Inland Church and Benedict
of the Palm Shade Hotel provided crucial logistical assistance in the Loiyangalani area. Jim Pierson
and Jeaneth Gomez at the Luminescence Dating
Research Laboratory at UIC ran the OSL samples
and made the analysis work when it seemed as if
it would not work. Junkyu Kim assisted with the
creation of the lithic figures. Finally, we would like
to thank our wives, Kristina Dziedzic Wright and
Liliana Marín who did the important work while we
were playing in the sandbox.
The two sites that were subject to subsurface testing as part of the project have temporal components that occur within or immediately preceding the time period spanning from
2500 to 1600 BP when terrigenous pollen spectra
from a deepwater sediment core reflect a climate
shift to arid conditions (Mohammed et al. 1995).
We presume that the settlements were close to
the lake margin, and people were opportunistically foraging when resources were available.
Conclusion
Preliminary research in the Porr strand plain
of southeastern Lake Turkana has identified 25 archaeological sites, primarily from the late Holocene, which indicate repeated, but ephemeral use
of the landform. Two archaeological sites (GeJk5,
GeJk28) located near the presumed Holocene
transgressive mark for Lake Turkana have denser
archaeological deposits, but lack bone harpoons
found in other assemblages adjacent to Galana Boi
sediments. Fauna recovered from surface contexts
and screened test units indicates hunting and fishing
activities continued in the area after 3000 BP. Diagnostic ceramics were primarily Nderit tradition,
but a previously unreported “corrugated” variant
was located at GeJk20, GeJk21, and GeJk22 during
pedestrian reconnaissance. The lithic assemblages
recovered are geared toward chopping, hunting,
cutting, boring, and scraping activities, but not fishing or plant cultivation or processing. Obsidian artifacts and the presence of Nderit tradition ceramics
show that the inhabitants of the Porr landform were
in contact with their neighbors to the west and north
Bibliography
Barthelme, J.
1977 Holocene sites north-east of Lake Turkana. 59
NYAME AKUMA Azania 12: 33-41.
2011 Four middle Holocene pillar sites in West Turkana, Kenya. Journal of Field Archaeology 36(3): 181-200.
Barthelme, J. W.
1985
Fisher-Hunters and Neolithic Pastoralists in East Turkana, Kenya. British Archaeo-
logical Reports International Series 254. Oxford: Archaeopress.
Lynch, B. M. and L. H. Robbins
1978 Namoratunga: The first archeoastronomical evidence in sub-Saharan Africa. Science 200(4343): 766-768.
Beyin, A.
2011
No. 76 DECEMBER 2011
1979 Cushitic and Nilotic prehistory: New
archaeological evidence from north-west Kenya. Journal of African History 20(3): 319-328.
Recent archaeological survey and excava-
tion around the greater Kalokol area, west side of Lake Turkana: Preliminary findings. Nyame Akuma 75: 40-50.
Marshall, F., K. Stewart and J. W. Barthelme
Bower, J. R. F. and C. M. Nelson
1984 Early domestic stock at Dongodien in
northern Kenya. Azania 19: 120-129.
1978 Early pottery and pastoral cultures of the
Central Rift Valley, Kenya. Man 13: 554-
566.
Mehlman, M. J.
1989
Brain, C. K.
1981 The Hunters or the Hunted? An Introduc-
tion to African Cave Taphonomy. Chicago:
University of Chicago Press.
Later Quaternary Archaeological Sequenc-
es in Northern Tanzania. PhD dissertation: University of Illinois at ChampaignUrbana.
Mohammed, M. U., R. Bonnefille and T. C. Johnson
Butzer, K. W.
1995 Pollen and isotopic records in Late Holo-
cene sediments from Lake Turkana, Kenya. Palaeogeography Palaeoclimatology
Palaeoecology 119: 371-383.
1980 The Holocene lake plain of north Rudolf, East Africa. Physical Geography 1: 42-58.
Clark, G.
Morgan, W.
1977 World Prehistory in New Perspective.
Cambridge: Cambridge University Press.
1981 Ethnobotany of the Turkana: Use of plants by a pastoral people and their livestock in Kenya. Economic Botany 35(1): 96-130.
Ehret, C.
2011 History and the Testimony of Language. Berkeley: University of California Press.
Nelson, C. M.
1973 A Comparative Analysis of Later Stone Age
Occurrence in East Africa. PhD disserta-
tion: University of California at Berkeley.
Hildebrand, E., J. Shea and K. Grillo
60
NYAME AKUMA 1995 Work of the Koobi Fora field school at the Jarigole Pillar Site. Kenya Past and Present 27: 49-63.
1984
Owen, R. B., J. W. Barthelme, R. W. Renaut and A.
Vincens
1982 Palaeolimnology and archaeology of Holo-
cene deposits north-east of Lake Turkana, Kenya. Nature 298: 523-529.
Late prehistoric and aquatic adaptations west of Lake Turkana, Kenya. In J. D. Clark and S. A. Brandt, editors, Hunters
to Farmers: Causes and Consequences of Food Production in Africa. Berkeley: Uni-
versity of California Press, pp. 206-211.
Robbins, L. H.
1972 Archeology in the Turkana District, Kenya. Science 176(4033): 359-366.
1974 The Lothagam Site. Anthropological Series,
Vol 1, No. 2. East Lansing: Michigan State University Museum.
Owen, R. B. and R. W. Renaut
1986 Sedimentology, stratigraphy and palaeoen-
vironments of the Holocene Galana Boi Formation, NE Lake Turkana, Kenya. Geo-
logical Society, London, Special Publica-
tions 25(1): 311-322.
1980
Phillipson, D. W.
Lopoy: A Late Stone Age Fishing and
Pastoralist Settlement in the Lake Turkana Basin, Kenya. Anthropological Series, Vol 3, No. 1. East Lansing: Michigan State
University Museum.
2006 Lake Turkana archaeology: The Holocene. Ethnohistory 53(1): 71-93.
1977Lowasera. Azania 7: 1-32.
1984
No. 76 DECEMBER 2011
Aspects of early food production in
northern Kenya. In L. Krzyzaniak and M. Kobusiewicz, editors, Origin and Early
Development of Food-Producing Cultures in North-Eastern Africa. Poznan: Polish Academy of Sciences, pp. 489-495.
Schlee, G.
1985 Interethnic clan identities among Cushitic-
speaking pastoralists. Africa 55(1): 17-38.
Soil Survey Staff
Reimer, P. J., M. J. L. Baillie, E. Bard, A. Bayliss,
J. W. Beck, P. G. Blackwell, C. Bronk Ramsey, C.
E. Buck, G. S. Burr, R. L. Edwards, M. Friedrich, P.
M. Grootes, T. P. Guilderson, I. Hajdas, T. J. Heaton, A. G. Hogg, K. A. Hughen, K. F. Kaiser, B.
Kromer, F. G. McCormac, S. W. Manning, R. W.
Reimer, D. A. Richards, J. R. Southon, S. Talamo,
C. S. M. Turney, J. van der Plicht and C. E. Weyhenmeyer
1993 Soil Survey Manual, no. 18. Washington D.C.: United States Department of Agricul-
ture.
1999
2010 INTCAL09 and MARINE09 Radiocarbon age calibration curves, 0–50,000 Years Cal BP. Radiocarbon 51(4): 1111-1150.
Soil Taxonomy: A Basic System of Soil Classification for Making and Interpreting Soil Surveys. 2nd ed. Washington, D. C.: United States Department of Agriculture, Natural Resources Conservation Service.
2002 Field Book for Describing and Sampling Soils. Washington, D.C.: National Soil Sur-
vey Center, Natural Resources Conserva-
tion Service, U.S. Department of Agricul-
ture.
Robbins, L.
61
NYAME AKUMA Spear, T. and R. Waller (editors)
1993 Being Maasai: Ethnicity and Identity in East Africa. London: James Currey.
Stave, J., G. Oba, I. Nordal and N. Stenseth
2007
Traditional Ecological Knowledge of a
Riverine Forest in Turkana, Kenya: Impli-
cations for Research and Management. Bio
diversity and Conservation 16(5): 1471-
1489.
62
No. 76 DECEMBER 2011